1 Introduction

In recent years, titanium alloys have become indispensable new structural materials due to the rapid development of aerospace and automotive fields (Ref 1). Titanium alloys possess good hardenability, excellent corrosion resistance, superior mechanical properties and high-strength-to-weight ratio (Ref 2, 3). However, many researchers are interested in the development of new titanium alloys due to the modifications of aircraft design criteria. In particular, the development of titanium alloys with high-specific strength and ductility is an attractive research direction (Ref 4). At present, a large number of high-strength titanium alloys have been developed via continuous efforts, including Ti-5553 (Ti-5Al-5Mo-5V-3Cr), Ti-15-3 (Ti-15V-3Cr-3Sn-3Al) and Ti-10-2-3 (Ti-10V-2Fe-3Al) alloys. These alloys are used in load-bearing components, elastic components, fasteners and hydraulic pipes in aviation aircraft and weapons equipment (Ref 5, 6). They are popular because their ultimate tensile strength exceeds 1250 MPa, and their fracture toughness is greater than 55 MPa m1/2 after solution treatment and aging (STA) (Ref 7, 8). However, the disadvantage of these high-strength titanium alloys is their excessive cost (Ref 9), which is due to the following two factors: (1) low thermal conductivity of titanium alloys leads to expensive processing; and (2) special processing methods are required to limit the content of interstitial elements such as oxygen, nitrogen and carbon (Ref 10).

Powder metallurgy (PM) has become a key technique to promote the development of titanium alloys with its advantages of low cost, no segregation and good comprehensive performance (Ref 11, 12). Many PM titanium alloys with excellent properties have been reported, such as Ti-6Al-4V, Ti-5Al-2.5Fe and Ti-6Al-1.7Fe-0.1Si alloys (Ref 13). In addition, many studies have employed solution treatment and aging to achieve excellent mechanical properties by adjusting α and β phases (Ref 14). For example, after solution treatment and aging, Ti-3Al-2.7Nb-15Mo (TB8) alloy achieved ultimate tensile strength of 1400 MPa and elongation of 8% (Ref 15). Phase transition temperature is the basis for formulating the material heat treatment process and selecting the hot working deformation parameters (Ref 16). However, due to the difference in chemical composition and processing techniques, (α + β)/β phase transition point is different for various titanium materials (Ref 17). Therefore, in order to control the microstructure and mechanical properties after solution treatment and aging, it is necessary to accurately determine the beta transus temperature of the material.

Ti-6554 (Ti-6Cr-5Mo-5V-4Al) is a metastable β titanium alloy with excellent mechanical properties, which has good prospects in aerospace applications (Ref 18). Ti-6554 alloy exhibits ultimate tensile strength of 1260 MPa and fracture toughness of 80 MPa m1/2 (Ref 19), and its mechanical properties are better compared to BT-22 and Ti-1023 alloys (Ref 7, 20). Hence, many researchers have paid much attention to Ti-6554 alloy. For example, Li et al. (Ref 21) studied the thermal deformation behavior of Ti-6554 alloy in the temperature range of 800-950 °C and strain rate range of 0.001-10 s−1 and analyzed its dynamic recrystallization mechanism. Kent et al. (Ref 22) studied the effects of aging temperature and heating rates on microstructure and hardness. It can be seen that these studies mostly focus on ingot metallurgy (IM) Ti-6554 alloys. There are almost no systematic studies on PM Ti-6554 alloys, which limits the development of PM high-strength titanium alloys.

In this work, low-cost PM Ti-6554 alloy was prepared using irregular powder via vacuum pressureless sintering followed by forging. In order to obtain high-performance, the forging billets were cooled by air cooling and water quenching, respectively. Beta transus temperature was studied by continuous heating. The effects of alloying elements and cooling rates on microstructure and mechanical properties were also evaluated. This research aims to obtain high-performance and low-cost PM Ti-6554 alloy, and the related results can provide theoretical support for microstructure evolution of PM Ti-6554 alloy.

2 Materials and Methods

2.1 Material Preparation

In this study, Ti-6554 alloy samples were prepared using hydrogenation and dehydrogenation (HDH) of titanium powder (99.9%, − 325 mesh) blended with alloying powders. Alloying powders including Al50-V50 (wt.%, \(<\) 50 μm), Cr (\(<\) 80 μm) and Mo (\(<\) 10 μm) were of low oxygen content (− 1500 ppm), which were purchased from Tiantailong (Tianjin) Metal Materials Co., Ltd. Their powder morphologies are shown in Fig. 1. All powders were mixed in an Ar gas airtight container (QXQM-2, Changsha Tencan Powder Technology Co., Ltd) for 3 h, using a roller mill at a speed of 160 rpm/min. Subsequently, the mixed powder was sealed into a silicone mold followed by pressing at 200 MPa for 60 s, using cold isostatic pressing (CIP) machine (Sichuan Suohaipu Isostatic Pressing Technology Equipment Co., Ltd). The inner cavity size of the silicone mold was 120 mm in diameter and 600 mm in height. After that, the green bodies were vacuum sintered in a vacuum sintering furnace (Shenyang Guangtai Vacuum Technology Co., Ltd.) at 1150 °C for 5 h under a vacuum of 10−2 Pa with a heating rate of 5 °C/min. Then, furnace cooling was carried out. For forging deformation, as-sintered billets were held at 1050 °C for 1 h in a chamber electric furnace. Then, the heated billets were immediately forged by an air forging hammer with deformation amount of 50%. Finally, as-forged samples were subjected to water quenching (WQ) and air cooling (AC), respectively. The schematic diagram of the preparation route is shown in Fig. 2. The chemical composition of Ti-6554 titanium alloy is listed in Table 1.

Fig. 1
figure 1

SEM images of (a) HDH titanium powders, (b) Al50V50 master alloy powders, (c) Cr powders and (d) Mo powders

Fig. 2
figure 2

Schematic diagrams showing preparation procedures for PM Ti-6554 alloy

Table 1 Chemical composition of Ti-6554 titanium alloy in this study

2.2 Beta Transus Temperature Observation and STA Process

The samples used for beta transus temperature observation were cut from as-forged alloys (WQ). First, the samples were cut into cylinders with diameter of 10 mm and height of 12 mm. Then, they were solution-treated at 770, 780, 790, 800C, 810 and 820 °C for 60 min, respectively, followed by water quenching.

According to the beta transus temperature of PM Ti-6554 alloy, as-forged alloys (WQ) were solution-treated at 800 °C for 30 min followed by furnace cooling. Then, aging treatment was carried out at 500 °C for 240 min followed by furnace cooling. The samples for STA treatment were cut into cylinders with diameter of 10 mm and height of 100 mm.

2.3 Sample Testing and Characterization

The interstitial O/N/H contents of raw powder and prepared samples were analyzed using an Eltra ONH-2000 machine. According to ASTM B328, the densities of as-sintered and as-forged samples were measured using Archimedes’ method. Relative densities were calculated by the ratio of measured density to theoretical density. Optical microscopy (OM, Olympus BX51), scanning electron microscopy (SEM, Hitachi SU6600) and electron backscattering diffraction (EBSD) were used to observe and characterize the microstructure and fracture morphology. In this study, the size of metallographic samples was φ10 × 12 mm2. The metallographic samples were prepared by sequential mechanical grinding, polishing and etching with Kroll’s reagent (volume proportion of HF:HNO3:H2O = 1:2:50). EBSD samples were mechanically ground with sand paper from 400# to 3000# and then subjected to suspension polishing to remove surface stress. X-ray diffraction analysis (XRD, Shimadzu XRD-6000) was performed (40 kV, 20 mA, Cu-Kα, 2°/min, 20°\(<\) 2θ\(<\) 80°) to identify the phase constitution. Tensile test was carried out at room temperature using an AGI-250KN machine, according to ASTM-E08 standard. The samples for tensile test were cut into M6 standard bars. Hardness test was performed on a TH320 Rockwell hardness machine, by applying 60 Kgf load for 10 s.

3 Results and Discussion

3.1 Beta Transus Temperature

Metastable β titanium alloys usually undergo transformation from beta to alpha phase during heat treatment. When solution temperature of titanium alloy is lower than the beta transus temperature, primary α phases are retained in the matrix and accumulated in the grain boundaries, thereby inhibiting the growth of β grains (Ref 23). Conversely, when solution temperature is higher than the beta transus temperature, β grains with large size are found in the alloy without primary α phases (Ref 23, 24). Therefore, the beta transus temperature is an important indicator to determine the heat treatment process. Moreover, it is reported that when the content of β phases exceeds 95%, the lowest quenching temperature is considered the beta transus temperature of titanium alloy (Ref 25). In this study, continuous heating was used to measure the beta transus temperature of PM Ti-6554 alloy. Figure 3 shows the microstructures of PM Ti-6554 alloys after holding at 770, 780, 790, 800, 810 and 820 °C, respectively. As shown in Fig. 3(a), (b), (c) and (d), there is a large amount of α precipitates in β matrix. However, the content of α precipitates gradually decreases with the increase in temperature. Moreover, it can be seen from Fig. 3(e) that the microstructure mainly consists of β phase with a small amount of α precipitates. The α precipitates almost disappear with the increase in temperature to 820 °C (Fig. 3f). For more accurate analysis, X-ray diffraction results are presented in Fig. 4, which are consistent with the preliminary results of this study. Therefore, the beta transus temperature of PM Ti-6554 alloy is about 810 \(\pm \) 10 °C.

Fig. 3
figure 3

Microstructure of PM Ti-6554 alloy after quenching at different temperatures of (a) 770, (b) 780, (c) 790, (d) 800, (e) 810 and (f) 820 °C

Fig. 4
figure 4

XRD analysis of PM Ti-6554 alloys after quenching at different temperatures

Chemical composition of titanium alloys can significantly affect the beta transus temperature of (α + β)/β titanium alloy, as shown in Table 2. Among them, molybdenum and vanadium have the same lattice type as β-Ti, which can be infinitely intersoluble in β titanium, thus reducing the (α + β)/β phase transition point to a certain extent (Ref 26, 27). Aluminum is inclined to dissolve in α-Ti and increases the beta transus temperature, while chromium forms continuous solid solution in β-Ti and reduces the beta transus temperature (Ref 27). Therefore, the mathematical calculation formula of beta transus temperature can be established according to the nominal composition of titanium alloys. According to Eq 1 (Ref 26), the theoretical beta transus temperature of PM Ti-6554 alloy is about 800 °C, which is in line with the measured value (810 \(\pm \) 10 °C). In Eq 1, 885 °C is the phase transus temperature of pure titanium. T(α+β)/β is the β transus temperature, M[Al/N/O/Mo/H/V/Cr/Mo] represents the mass ratio of each element, and ΔTβ represents the element impact on (α + β)/β phase transition.

$$ \begin{aligned} T_{{\left( {\alpha + \beta } \right)/\beta }} =\, & 885 + M_{{\text{[Al]}}} \times \Delta T_{{\beta [{\text{Al}}]}} + M_{{\text{[N]}}} \times \Delta T_{{\beta [{\text{N}}]}} \\ & + M_{{[{\text{O}}]}} \times \Delta T_{{\beta [{\text{O}}]}} + M_{{[{\text{Mo}}]}} \times \Delta T_{{\beta [{\text{Mo}}]}} + M_{{[{\text{H}}]}} \\ & \times \Delta T_{{\beta [{\text{H}}]}} + M_{{[{\text{V}}]}} \times \Delta T_{{\beta [{\text{V}}]}} + \, M_{{[{\text{Cr}}]}} \times \Delta T_{{\beta [{\text{Cr}}]}} \\ \end{aligned} $$
(1)
Table 2 Effect of element content on (\(\alpha +\beta \))/β phase transition point (Ref 26)

3.2 Phase and Microstructure Analysis

Figure 5 shows XRD patterns of PM Ti-6554 alloys in different states. It can be seen that the main peak intensity (relative intensity) of as-forged alloy is much higher than that of as-sintered alloy, while α phases gradually disappear. Moreover, the main peak intensity of as-forged (WQ) alloy is higher than that of as-forged (AC) alloy. This indicates that as-forged Ti-6554 alloy is mainly composed of β phases, and the volume fraction of β phases increases with the increase in cooling rate. Jones et al. (Ref 28) found that water quenching with fast cooling rate can effectively prevent the precipitation of α phases from β phase. This result is the same as that for PM Ti-6554 alloys. It can also be seen that a large number of α phases are produced in β matrix after solution treatment and aging. Moreover, Burgers orientation relationship exists between α and β phases. Normally, the distribution of plastic flow is different between bcc β phase and hcp α phase. Bcc β phase has a lower flow stress compared to hcp α phase. Therefore, β grains tend to exhibit high yield strength due to the overall weaker weave, which does not have large shear stresses inside (Ref 29). Furthermore, microstructure evolution during deformation mainly depends on bcc β phase, which needs to be studied further with other methods.

Fig. 5
figure 5

XRD analysis of PM Ti-6554 alloys under sintering, forging and STA conditions

Figure 6(a) shows the microstructure of as-sintered Ti-6554 alloy. It can be seen that a large number of α phases are aggregated at the grain boundary of β phase. However, the number of α phases within β grains is significantly reduced. This indicates that α phases are more likely to precipitate at the β phase grain boundaries. Ohmori et al. (Ref 30) also confirmed that α phase nucleates preferentially at the β phase grain boundaries. The microstructures of as-forged samples are distinctly different from those of as-sintered samples, as shown in Fig. 6(b) and (c). Note that the cooling rate also has an effect on the microstructures. After forging deformation, some β grains are recrystallized and refined, while other β grains grow up due to the aggregation of deformation. More importantly, the content of α phase precipitates decreases significantly after air cooling. It has been reported that when the cooling rate is high, β phase grains can nucleate inside and form independent α sheets. When the cooling rate is slow, β phase grains become clusters, and α slices grow from the grain boundary to the grain interior throughout the whole grain. However, this kind of microstructure is often not conducive to the improvement of elongation (Ref 30). As shown in Fig. 6(c), the water-quenched microstructure after forging is completely β phase structure, and α phase fully disappears.

Fig. 6
figure 6

Microstructures of Ti-6554 alloy in different states: (a) as-sintered, (b) as-forged with air cooling, (c) as-forged with water quenching and (d) STA

In the process of solution treatment and aging, the morphology and number of α phases depend on the aging temperature and cooling rate, while the grain size of β phase depends on the solution temperature. Figure 6(d) shows the microstructure of STA sample. Many α phases are uniformly dispersed in the β matrix. A large number of lamellar α phases are precipitated on the β phase grain boundary, whose thickness is about 0.5-1 μm. Moreover, thin lamellar α phases are precipitated on the β matrix, whose thickness is less than 0.5 μm. This indicates that α phase first precipitates at the grain boundary of the β phase during the solution treatment and aging. With increase in holding time, the acicular \(\alpha \) phase at the grain boundary gradually grows into lamellar shape, and the thickness gradually increases.

Grain texture of Ti-6554 alloy was analyzed by EBSD characterization, as shown in Fig. 7. Low-angle grain boundaries (LAGBs) and high-angle grain boundaries (HAGBs) are distinguished by different colors. The thin white line is the LAGB with an orientation deviation of 2-15°, and the thick black line is the HAGB with an orientation deviation of > 15°. Figure 7(a1), (b1) and (c1) shows the inverse pole figures (IPF) of PM Ti-6554 alloy at different states. Different colors of grains in the figure indicate different grain orientations. It can be seen that there is no preferred orientation in the three alloys. As shown in Fig. 7(d), there is a large amount of α phase precipitates, of which − 40% α phases are rodlike secondary α with grain size of 2-6 μm, and − 60% α phases are acicular secondary α with grain size of − 2 μm. Figure 7(a2), (b2) and (c2) shows the corresponding pole figures of PM Ti-6554 alloy in {100} direction. The results indicate that the grains have no specific orientation.

Fig. 7
figure 7

IPF mapping and {100} pole figure of PM Ti-6554 titanium alloy: (a1, a2) forging with air cooling, (b1, b2) forging with water quenching and (c1, c2, d) STA

The grain size distribution and misorientation angle difference of as-forged Ti-6554 alloy under air cooling and water quenching conditions are presented in Fig. 8(a1), (a2), (a3), (b1) and (b2). It can be seen that after air cooling, LAGB accounts for 85% of all grain boundaries, while it is about 70% in the water-quenched sample. In both samples, HAGB accounts for a very small proportion. This may be because very few grains are dynamically recrystallized during forging in PM Ti alloy. After forging with air cooling, the average grain size of β phase is about 11.04 μm, which is higher than that of water-quenched alloy.

Fig. 8
figure 8

Grain size distribution and misorientation angle distribution of PM Ti-6554 titanium alloy: (a1, a2) forging with air cooling, (b1, b2) forging with water quenching and (c1, c2) STA

To regulate the mechanical performance, the water-quenched samples were solution-treated at 800 °C for 30 min followed by aging heat treatment at 500 °C for 4 h. The average grain size of β grain is about 2.3 μm. Furthermore, LAGB accounts for 96% of all grain boundaries, as shown in Fig. 8(c2). In the hot deformation process, dynamic recovery and dynamic recrystallization occur, resulting in the increase in storage distortion energy, dislocation density and LAGB. Consequently, the alloy strength increases at the expense of plasticity.

3.3 Mechanical Performance

Table 3 shows the room temperature tensile properties and relative densities of PM Ti-6554 alloys in different states. According to Table 3 and Fig. 9, the relative density (RD) of as-sintered Ti-6554 alloy is 93%. This result is consistent with the microstructure containing some residual pores, as shown in Fig. 6(a). Nevertheless, the RD of as-forged samples increases up to 99.9%, and microstructure is almost fully dense. Compared with as-sintered samples, the ultimate tensile strength (UTS), yield strength (YS) and elongation (EL.) of as-forged samples are greatly improved. This result can be explained by two reasons. First, β and α grains are broken into small size with deformation, which can improve ultimate tensile strength and yield strength. Second, residual pores are eliminated, which can significantly improve the elongation. It is worth noting that the elongation of as-forged (AC) alloy is only 4%, while it is 20% for as-forged (WQ) alloy, with an increase of 16%. This is mainly because microstructure of as-forged (WQ) alloy is composed of β phase, as shown in Fig. 6(c). Compared with α phase, β phase has more slip systems with body-centered cubic structure, which is conducive to dislocation movement during plastic deformation (Ref 31). However, α phase has a dense hexagonal structure with fewer slip systems, which has a strengthening effect at the expense of ductility (Ref 31). Thus, as-forged (WQ) Ti-6554 alloy exhibits excellent mechanical properties of 942 MPa (UTS), 888 MPa (YS) and 10% (EL.). Finally, PM Ti-6554 titanium alloy shows remarkable ultimate tensile strength (1211 MPa) and good elongation (13.8%), which is a good balance between strength and ductility.

Table 3 Properties of PM TI-6554 titanium alloys
Fig. 9
figure 9

Stress–strain curves of PM Ti-6554 titanium alloys

3.4 Fracture Morphology

Fracture morphologies of PM Ti-6554 alloys are shown in Fig. 10. As seen from Fig. 10(a), many intergranular cracks are formed at the interface of β grains. Although there are small dimples distributed in some regions, cleavage facets can be clearly observed. Thus, it is a typical brittle fracture mode. This is mainly because residual pores are not eliminated, resulting in crystal defects such as holes and dislocations in the grains, which lead to blocked expansion of slip band and uneven deformation during the tensile test (Ref 32). In such case, stress concentration occurs and brittle fracture with uneven surface is formed. The fractography of as-forged (AC) samples is shown in Fig. 10(b). It can be seen that there are many small and shallow dimples in the fracture surface. The ductility of this alloy is usually determined by the depth and size of dimples. When the dimples are deep and large, the alloy has good ductility. Conversely, if the dimples are shallow and small, the alloy has poor ductility (Ref 33). The path of some cracks spreading along the intergranular fracture can also be seen. This is mainly due to the precipitation of brittle α phase at β phase boundary in these regions, which leads to the decrease in toughness and intergranular brittle fracture. The fractography of as-forged (WQ) sample is shown in Fig. 10(c). The characteristics of river pattern and cleavage facets are present, the number of small dimples is reduced, and the fracture toughness of the alloy is increased. These observations are in agreement with the tensile test results in Fig. 9. Figure 10(d) presents the fractography of STA sample, which shows the distribution of obvious dimples with few intragranular cracks.

Fig. 10
figure 10

Typical fracture images of PM Ti-6554 titanium alloys: (a) as-sintered, (b) forging with air cooling, (c) forging with water quenching, and (d) STA

4 Discussion

4.1 Effects of Alloying Elements

Generally, titanium alloys are made up of beta stabilizers (Mo, V, Fe, Co, Nb, Cu), alpha stabilizers (N, C, O, Al, B) and neutral elements (Zr, Sn) (Ref 26). Among them, alpha stabilizers can significantly improve the temperature of stable region of alpha phase. Conversely, beta stabilizers reduce the stable region temperature of beta phase (Ref 25). In addition, beta and alpha stabilizers also influence the molybdenum equivalent ([Mo]eq) and aluminum equivalent ([Al]eq), which can be described using Eq 2 and 3 (Ref 34). Previous research has proved that molybdenum equivalent and aluminum equivalent can affect the stability, phase composition and beta transus temperature of titanium alloys (Ref 35). Moreover, molybdenum equivalent and aluminum equivalent have a great effect on ductility, strength and fracture toughness (Ref 36). I. Weiss et al. (Ref 37) found that beta transus temperature reduces with the increase in molybdenum equivalent. Thus, titanium alloys have a wide hot working window. According to Eqs. (2) and (3) and Table 1, PM Ti-6554 alloy has molybdenum equivalent of 18.7 and aluminum equivalent of 7.17, which can be classified as metastable β titanium alloy. Therefore, PM Ti-6554 alloy can achieve excellent strength over a wide hot working window.

$$ \left[ {{\text{Mo}}} \right]_{{{\text{eq}}}} = \% {\text{Mo}} + \% {\text{V}}/{1}.{5} + \% {\text{Cr}}/0.{63} $$
(2)
$$ \left[ {{\text{Al}}} \right]_{{{\text{eq}}}} = \% {\text{Al}} + {1}0\% \left[ {\text{O}} \right], \, \left( {\left[ {\text{O}} \right] = {\text{O}} + {\text{2N}}} \right) $$
(3)

In metastable β-Ti alloys, alloying elements also have a great influence on phase formation and microstructure morphology. A number of studies have found that α stabilizers and β stabilizers have different diffusion rates (Ref 17). In the heating process of titanium alloy, partial polymerization of α stabilizers leads to α phases, and the aggregation of β stabilizers forms β phases (Ref 38). Therefore, in the forging heating and cooling process, different elements diffuse at different rates. When as-forged samples are cooled by air, the slow cooling rate allows sufficient time for diffusion to occur, resulting in the precipitation of secondary α phases in the β grain boundary. However, when as-forged samples are cooled by water, α stabilizer does not have enough time for partial polymerization. Consequently, the microstructures are composed of β phases without α phase precipitates.

4.2 Effects of Cooling Rates

As is well known, the microstructure of titanium alloy has a significant impact on its mechanical properties. The strength and ductility of as-forged samples exhibit large differences due to different cooling rates, as shown in Table 3. It can be clearly seen from Fig. 8(a1) and (b1) that the average grain size of the β phase of as-forged (AC) samples (− 11.38 μm) is almost the same as that of as-forged (WQ) samples (− 10.03 μm). However, large β phase (10-40 μm) in as-forged (AC) samples accounts for a greater proportion, and some α phases are precipitated in the β matrix. As-forged (AC) samples consist of β phases without α phases, as shown in Fig. 6(c). Hence, PM Ti-6554 alloy displays good mechanical properties (942 MPa tensile strength and 20% elongation). After solution treatment and aging, PM Ti-6554 titanium alloy exhibits fine grain size of 2.3 μm and uniform α phase precipitates in β phase, including 40% rod-like α and 60% acicular secondary α. Overall, Ti-6554 alloy possesses a good balance between strength and plasticity, with ultimate tensile strength of 1211 MPa and elongation of 13.8%.

Comparison of the mechanical properties of some typical high-strength titanium alloys is shown in Table 4. It can be seen that the previously reported Ti-6554 alloys are mainly prepared by traditional ingot metallurgy technology. PM Ti-6554 alloys have better mechanical properties overall. Compared to Ref.39, higher strength and plasticity are obtained in PM Ti alloy, indicating that PM technology is a promising method to prepare high-strength Ti alloys. In addition, strength improvement always results in the reduced elongation in metastable β titanium alloys, which is basically in line with the results shown in Table 4. Moreover, PM Ti-6554 alloy has better mechanical properties compared with other PM high-strength titanium alloys, such as PM Ti-5553 and PM Ti-1023 alloys. Through the optimization of heat treatment parameters, titanium alloys can achieve a good balance between strength and plasticity. Therefore, a reasonable heat treatment process will be studied to prepare PM Ti-6554 alloy with higher performance in the follow-up experimental work.

Table 4 Tensile properties of some typical high-strength titanium alloys

5 Conclusion

In this paper, the microstructure evolution and mechanical properties of PM Ti-6554 alloy in different states were systematically studied. The effects of alloying elements and cooling rate on microstructure and mechanical properties were also investigated. The main findings are as follows:

  1. (1)

    Ultra-high-strength Ti-6554 alloy prepared using irregular powder via vacuum pressureless sintering and forging. PM Ti-6554 alloy achieved a fully dense microstructure without segregation of components.

  2. (2)

    As-forged (WQ) Ti-6554 alloy consisted of β phases, whose average grain size was 10 μm. The alloy achieved ultimate tensile strength of 942 MPa, yield strength of 888 MPa and elongation of 20%. Moreover, PM Ti-6554 alloy possessed a good balance between strength and plasticity (UTS of 1211 MPa, YS of 1201 MPa and EL. of 13.8%) with fine α phase (< 2 μm) precipitates in β phase.

  3. (3)

    Cooling rate had a significant impact on microstructure and mechanical properties. At faster cooling rate, there was insufficient time for α phase to precipitate in the β matrix. Consequently, the as-forged PM Ti-6554 alloy possessed good mechanical properties.